javax.datamining.data
Interface PhysicalDataSet

All Superinterfaces:

MiningObject

public interface PhysicalDataSet

extends MiningObject

The interface PhysicalDataSet serves as a common interface for describing data to be used as input to data mining. Through the mapping of attributes, attributes of the physical data are mapped to logical attributes of a model?s logical data. The data referenced by a physical data object can be used in, for example, model building, model application (scoring), and test.

The physical attributes are ordered and can be accessed individually by index, which ranges between 0 and n-1, where n is the number of physical attributes.

Author:

JSR-73 Java Data Mining Expert Group

See Also:

MiningObject

Method Summary

void	addAttribute(PhysicalAttribute attribute) Adds a physical attribute to the physical data.
void	addAttributes(PhysicalAttribute[] attributeArray) Adds an array of physical attributes to the physical data.
PhysicalAttribute	getAttribute(int index) Returns the PhysicalAttribute that corresponds to the specified index.
PhysicalAttribute	getAttribute(java.lang.String attributeName) Returns an instance of PhyisicalAttribute with the specified name present in the physical data.
int	getAttributeCount() Returns the number of PhysicalAttribute objects contained.
java.lang.Integer	getAttributeIndex(java.lang.String attributeName) Returns the index of the specified PhysicalAttribute name.
java.util.Collection	getAttributes() Returns a collection of PhyisicalAttribute instances present in the physical data.
AttributeStatisticsSet	getAttributeStatistics() Returns the AttributeStatisticsSet instance if any statistics were computed for the PhysicalData.
java.lang.String	getURI() Returns the URI of the physical data.
void	importMetaData() Imports the metadata of the physical data to populate the physical attributes.
void	removeAllAttributes() Removes all physical attributes currently associated with the PhysicalDataSet instance.
void	removeAttribute(java.lang.String name) Removes the specified attribute from the physical data.

Methods inherited from interface javax.datamining.MiningObject

getCreationDate, getCreatorInfo, getDescription, getName, getObjectIdentifier, getObjectType, setDescription

Method Detail

addAttribute

public void addAttribute(PhysicalAttribute attribute)
                  throws JDMException

Adds a physical attribute to the physical data. The attribute is appended to the end of the list.

Parameters:

attribute - A physical attribute to be added.

Returns:

void

Throws:

JDMException

addAttributes

public void addAttributes(PhysicalAttribute[] attributeArray)
                   throws JDMException

Adds an array of physical attributes to the physical data. The attributes are appended to the list.

Parameters:

attributeArray - An array of physical attributes to be added.

Returns:

void

Throws:

JDMException

getAttribute

public PhysicalAttribute getAttribute(int index)

Returns the PhysicalAttribute that corresponds to the specified index.

Parameters:

index - The index of the physical attribute to be returned.

Returns:

PhysicalAttribute

getAttribute

public PhysicalAttribute getAttribute(java.lang.String attributeName)
                               throws JDMException

Returns an instance of PhyisicalAttribute with the specified name present in the physical data.

Parameters:

attributeName - The name of the physical attribute to be returned.

Returns:

PhysicalAttribute

Throws:

JDMException

getAttributeCount

public int getAttributeCount()

Returns the number of PhysicalAttribute objects contained.

Returns:

int

getAttributeIndex

public java.lang.Integer getAttributeIndex(java.lang.String attributeName)
                                    throws JDMException

Returns the index of the specified PhysicalAttribute name.

Parameters:

attributeName - The name of the physical attribute whose index is to be returned.

Returns:

Integer

Throws:

JDMException

getAttributes

public java.util.Collection getAttributes()
                                   throws JDMException

Returns a collection of PhyisicalAttribute instances present in the physical data.

The attributes are ordered by the index, i.e., the order by which each item is put into PhysicalDataSet. The first attribute in the collection is the item that was put first.

Returns:

Collection

Throws:

JDMException

getAttributeStatistics

public AttributeStatisticsSet getAttributeStatistics()
                                              throws JDMException

Returns the AttributeStatisticsSet instance if any statistics were computed for the PhysicalData. Returns null if the data statistics is not available.

Returns:

AttributeStatisticsSet

Throws:

JDMException

getURI

public java.lang.String getURI()

Returns the URI of the physical data.

Returns:

String

importMetaData

public void importMetaData()
                    throws JDMException

Imports the metadata of the physical data to populate the physical attributes. After import, the roles are all initialized to data, which can be overridden later. Metadata import provides a snapshot of the metadata at the time of invocation.

Invocation of this method effectively removes all physical attributes that are currently present in the object. If this method is not invoked and no physical attribute is added manually, i.e., the physical data set is empty, then the DME tries to obtain the metadata from the physical data specified by the URI when this physical data set is used.

Statistics may be populated automatically if they are available, this is vendor-specific. Statistics at this level may be approximate or exact.

If the data contains attributes that are not string, double, or integer, the vendor determines the mapping from the data source type to the JDM attribute type.

If a data type is not supported or mappable to one of the standard types, it has attribute data type of unknown.

Returns:

void

Throws:

JDMException

removeAllAttributes

public void removeAllAttributes()

Removes all physical attributes currently associated with the PhysicalDataSet instance.

Returns:

void

removeAttribute

public void removeAttribute(java.lang.String name)
                     throws JDMException

Removes the specified attribute from the physical data. The remaining attributes are shifted towards the beginning of the list, i.e., the index values are decreased.

If name not found, an exception is thrown.

Parameters:

name - The name of the physical attribute to be removed.

Returns:

void

Throws:

JDMException